Introduction. The selectins are a family of three adhesion molecules (L-, E- and P-selectin) that are involved in the initial attachment of blood-borne leukocytes to endothelial cells during the process of emigration from the bloodstream into the surrounding tissue. All three selectins bind to carbohydrate-based ligands on opposing cells in a calcium-dependent manner. L-selectin is unique among the selectins by virtue of its constitutive expression on all classes of circulating leukocytes. In addition, L-selectin plays a key role in leukocyte recruitment during a number of acute and chronic inflammatory conditions, focusing a tremendous amount of interest on the nature of the carbohydrate ligands on opposing endothelial cells. We have initiated a program aimed at the structural identification of carbohydrate ligands for L-selectin. Our approach involves analysis of the oligosaccharide structures on biological selectin ligands and the chemical synthesis of identified structures to directly demonstrate functional activity. Mass spectrometry is central to the characterization of our synthetic products, and will be the principal analytical tool in the direct structural identification of the carbohydrate epitopes on biological L-selectin ligands. Results and Discussion. Previous work in this laboratory has led to the molecular identification of two biological glycoprotein ligands for L-selectin, termed GlyCAM-1 and CD34. The oligosaccharides on these glycoproteins are sulfated and sialylated, two modifications which were shown to be essential for L-selectin recognition. Preliminary characterization of the oligosaccharides on GlyCAM-1 using metabolic radiolabeling techniques has revelealed the presence of a novel capping group, 6'-sulfo sialyl Lewis x [NeuAca2,3(SO4-6)Galb1,4(Fuca1,3)GlcNAc, 1]. Thus, it is hypothesized that sulfation of the sialyl Lewis x tetrasaccharide on the 6'-position imparts high affinity binding activity to L-selectin. To test this hypothesis, we have designed a chemical/enzymatic synthesis for sulfated oligosaccharides related to structure 1. Our first target is compound 6 (scheme 1), in which the sialic acid residue of structure 1 has been replaced with a synthetically more accessible sulfate ester. The synthetic route begins with selective protection of the readily available disaccharide lactose (2) to afford derivative 3 in three steps. The 3'-, 4'- and 6'-positions are then selectively liberated with acid to afford compound 4. Chemical sulfation proceeds selectively at the 3'- and 6'-positions yielding, after deprotection, disulfated intermediate 5. The structures of intermediates 2-5 have been assigned in part using mass spectrometry. Finally, enzymatic fucosylation using a recombinant fucosyltransferase (FucT V) and GDP-fucose will afford target moledule 6. Currently, we have completed the synthesis of 5 and tested this intermediate for L-selectin binding activity. Preliminary results indicate that compound 5 binds to L-selectin more potently than similar derivatives lacking the sulfate ester at the 6'-position. Thus, this key sulfate ester appears to contribute significantly to L-selecting binding activity. We anticipate that synthetic oligosaccharides such as compound 6 will be even more potent as L-selectin antagonists, and may demonstrate anti-inflammatory activity in vivo. Finally, we plan to complement our metabolic radiolabeling analysis of the GlyCAM-1 oligosaccharides with direct characterizat ion by mass spectrometry. Towards this goal,